Random Forests for Time Series

Detalhes bibliográficos
Autor(a) principal: Goehry, Benjamin
Data de Publicação: 2023
Outros Autores: Yan , Hui, Goude , Yannig, Massart , Pascal, Poggi , Jean-Michel
Tipo de documento: Artigo
Idioma: eng
Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
Texto Completo: https://doi.org/10.57805/revstat.v21i2.400
Resumo: Random forests are a powerful learning algorithm. However, when dealing with time series, the time-dependent structure is lost, assuming the observations are independent. We propose some variants of random forests for time series. The idea is to replace standard bootstrap with a dependent block bootstrap to subsample time series during tree construction. We present numerical experiments on electricity load forecasting. The first, at a disaggregated level and the second at a national level focusing on atypical periods. For both, we explore a heuristic for the choice of the block size. Additional experiments with generic time series data are also available.
id RCAP_0ecb189ebe150d40eb5ae8788f62037f
oai_identifier_str oai:revstat:article/400
network_acronym_str RCAP
network_name_str Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
repository_id_str 7160
spelling Random Forests for Time SeriesBlock bootstrapRandom forestsRegressionTime seriesRandom forests are a powerful learning algorithm. However, when dealing with time series, the time-dependent structure is lost, assuming the observations are independent. We propose some variants of random forests for time series. The idea is to replace standard bootstrap with a dependent block bootstrap to subsample time series during tree construction. We present numerical experiments on electricity load forecasting. The first, at a disaggregated level and the second at a national level focusing on atypical periods. For both, we explore a heuristic for the choice of the block size. Additional experiments with generic time series data are also available.Statistics Portugal2023-06-26info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articleapplication/pdfhttps://doi.org/10.57805/revstat.v21i2.400https://doi.org/10.57805/revstat.v21i2.400REVSTAT-Statistical Journal; Vol. 21 No. 2 (2023): REVSTAT-Statistical Journal; 283–302REVSTAT; Vol. 21 N.º 2 (2023): REVSTAT-Statistical Journal; 283–3022183-03711645-6726reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAPenghttps://revstat.ine.pt/index.php/REVSTAT/article/view/400https://revstat.ine.pt/index.php/REVSTAT/article/view/400/643Copyright (c) 2021 REVSTAT-Statistical Journalinfo:eu-repo/semantics/openAccessGoehry, BenjaminYan , HuiGoude , YannigMassart , PascalPoggi , Jean-Michel2023-07-01T06:30:12Zoai:revstat:article/400Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-19T18:02:13.234448Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse
dc.title.none.fl_str_mv Random Forests for Time Series
title Random Forests for Time Series
spellingShingle Random Forests for Time Series
Goehry, Benjamin
Block bootstrap
Random forests
Regression
Time series
title_short Random Forests for Time Series
title_full Random Forests for Time Series
title_fullStr Random Forests for Time Series
title_full_unstemmed Random Forests for Time Series
title_sort Random Forests for Time Series
author Goehry, Benjamin
author_facet Goehry, Benjamin
Yan , Hui
Goude , Yannig
Massart , Pascal
Poggi , Jean-Michel
author_role author
author2 Yan , Hui
Goude , Yannig
Massart , Pascal
Poggi , Jean-Michel
author2_role author
author
author
author
dc.contributor.author.fl_str_mv Goehry, Benjamin
Yan , Hui
Goude , Yannig
Massart , Pascal
Poggi , Jean-Michel
dc.subject.por.fl_str_mv Block bootstrap
Random forests
Regression
Time series
topic Block bootstrap
Random forests
Regression
Time series
description Random forests are a powerful learning algorithm. However, when dealing with time series, the time-dependent structure is lost, assuming the observations are independent. We propose some variants of random forests for time series. The idea is to replace standard bootstrap with a dependent block bootstrap to subsample time series during tree construction. We present numerical experiments on electricity load forecasting. The first, at a disaggregated level and the second at a national level focusing on atypical periods. For both, we explore a heuristic for the choice of the block size. Additional experiments with generic time series data are also available.
publishDate 2023
dc.date.none.fl_str_mv 2023-06-26
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/article
format article
status_str publishedVersion
dc.identifier.uri.fl_str_mv https://doi.org/10.57805/revstat.v21i2.400
https://doi.org/10.57805/revstat.v21i2.400
url https://doi.org/10.57805/revstat.v21i2.400
dc.language.iso.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv https://revstat.ine.pt/index.php/REVSTAT/article/view/400
https://revstat.ine.pt/index.php/REVSTAT/article/view/400/643
dc.rights.driver.fl_str_mv Copyright (c) 2021 REVSTAT-Statistical Journal
info:eu-repo/semantics/openAccess
rights_invalid_str_mv Copyright (c) 2021 REVSTAT-Statistical Journal
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Statistics Portugal
publisher.none.fl_str_mv Statistics Portugal
dc.source.none.fl_str_mv REVSTAT-Statistical Journal; Vol. 21 No. 2 (2023): REVSTAT-Statistical Journal; 283–302
REVSTAT; Vol. 21 N.º 2 (2023): REVSTAT-Statistical Journal; 283–302
2183-0371
1645-6726
reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
instacron:RCAAP
instname_str Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
instacron_str RCAAP
institution RCAAP
reponame_str Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
collection Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
repository.name.fl_str_mv Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
repository.mail.fl_str_mv
_version_ 1799131688423391232